Abstract Although significant progress has been made in reducing smoking rates in the United States, one in seven Americans continues to smoke, and smoking remains the leading preventable cause of death in this country. Evidence suggests that tobacco control policies, including cigarette excise taxes, smokefree air laws and tobacco control program funding, can effectively reduce tobacco consumption in general. What remains unknown are which combinations of tobacco control policies are most effective, whether policy-related gains can be maintained during periods of policy retrenchment, and whether tobacco policies can shrink strong socioeconomic smoking disparities. To examine these questions, we propose to develop and analyze a dataset that integrates longitudinal observations of individual smoking behavior with measures of state and local tobacco control policy exposure, as well as area measures of socioeconomic status. Using repeated observations of 18,407 individuals' smoking behavior as policies change in their home state, county or city, or as they move to a new place with different policies, we aim to identify whether changes in adult smoking behavior following exposure to stronger tobacco control policies are the inverse of changes in behavior observed following exposure to weaker policies, and assess which policy combinations are most likely to be effective in the current context. By incorporating measures of individual and area income and educational status, we further aim to estimate the impact of single and combined tobacco control policy options on socioeconomic disparities in smoking. Our team has expertise in several areas, including tobacco control, policy evaluation and longitudinal data analysis. This innovative policy analysis method could be used to complement more traditional policy evaluation techniques for other health-related policies as well.